the associated press/Los angeLes times, ifran Khan, pooL

Positive Student Outcomes in Community Schools
Sebastian Castrechini and Rebecca A. London February 2012

w w

Positive Student Outcomes in Community Schools
Sebastian Castrechini and Rebecca A. London February 2012


1 Introduction and summary 5 Background on community schools 9 Redwood City context and research design 15 Study findings 23 Implications for policy, practice, and future research 33 Conclusion 35 Appendix 37 About the authors 38 Acknowledgements 39 Endnotes

Introduction and summary
In a nation where 42 percent of children live in low-income families, too many schools face the challenge of teaching students burdened with unmet needs that pose obstacles to learning. If our aim as a country is to ensure that all children succeed academically, particularly those living in struggling communities with limited resources, we simply can’t ask schools to do it alone. Community schools that align schools and community resources are a promising strategy for improving student outcomes by providing wraparound services that meet the social, physical, cognitive, and economic needs of both students and families. And while much of the current literature on community schools focuses on highlighting policies and practices to support the implementation of community school models, very little research examines how community schools affect student outcomes. Since 2007, the John W. Gardner Center for Youth and Their Communities ( JGC) at Stanford University has partnered with the Redwood City School District (RCSD) in Redwood City, Calif., south of San Francisco, to conduct research on participation and outcomes for students in the Redwood City School District’s community schools. This local initiative includes five community schools, with students in grades K through 8, that provided more than 250 programs, services, and events in the 2010-11 school year. The purpose of this paper is to provide an in-depth analysis of one district’s community schools using quantitative data to show how students and families use services at these schools and how those services work together to positively affect student outcomes. This analysis uses the Youth Data Archive, a JGC initiative that matches data across agencies that serve youth in common to ask and answer questions that the agencies could not answer alone. For this analysis, we linked student achievement data from the Redwood City School District, attendance records from program providers at community schools, and student survey data collected by the JGC, to examine participation patterns in community school programs as well as the relationship between these services and student outcomes.

introduction and summary |


The main findings from this analysis are: • Supplemental programs provided at Redwood City School District community schools reached more than 70 percent of the students enrolled at those schools. These programs generally served the most socioeconomically disadvan-

taged students, including high rates of students who were English learners, were eligible for subsidized meals, and had parents who had not completed high school. • English learner students with consistent program participation over time showed gains in English language development scores. In the elementary grades, these gains were tied to family engagement participation, but continued gains during middle school were associated with frequent extended learning program participation. • Community school programs were linked to positive attitudes about school for middle school students. Students with family engagement in elementary school entered middle school more likely to say that their school provided a supportive environment compared to those without family engagement. Once in middle school, frequent participation in extended learning programs was linked to increases in students’ perceptions of their school as a supportive environment. Feeling supported at school was linked to students’ motivation and academic confidence, both of which were associated with gains in achievement in math for all students and English language development scores for English learners. The findings have important implications for policy at the state, federal, and local levels. Key implications outlined in this paper include: • Expanding community schools to reach more students. Findings from this analysis show that community schools are a promising strategy for improving student outcomes. In the current climate of budget cuts, it is important for policymakers at the state and federal levels to advocate for community schools as an efficient, effective way to use scarce resources by leveraging partnerships. Districts can further help by creating district-wide community school initiatives. • Supporting improved outcomes for English learners at community schools. Academic score gains linked to community school program participation were especially strong for English learners in our analysis and were tied to receiving multiple services. Therefore, it is important for policymakers to promote collaborative structures at community schools in which students and families


center for american progress | positive student outcomes in community schools

are more likely to make use of the multiple services available. One method is to streamline the multiple and fractured funding sources that go into community schools to foster collaboration and common goals instead of competition for resources and disparate data collection and reporting. Findings in this analysis show that family engagement plays a key role, and local leaders can encourage family engagement by reaching out to parents and inviting them to be partners in a variety of different opportunities both at school and at home. • Fostering positive attitudes about school as a strategy for raising achievement. This analysis shows the critical role of student attitudes toward school and learning as one mechanism through which community schools affect achievement. The practices that foster these attitudes are well established in existing research, so it is important to ensure that all staff that interact with youth at community schools use these practices. Policymakers can adopt school climate standards and invest in measuring progress toward those standards, and local leaders can integrate these practices into existing classroom or program observation rubrics. • Building capacity for collecting and using data. Although data collection and analysis on the broad array of services offered at community schools is difficult, it is critical to informing policy and programmatic decision making. This involves sharing data among the many partners that provide services at community schools, and policymakers can make the process easier by clarifying and aligning regulations on data sharing at the federal and state levels. Local community school leaders can further help by developing shared goals and indicators among partners and creating a culture of sharing and examining data together with partners. Researchers are an important resource for community schools to involve in these efforts, and it will be important for the research field to find new strategies to meet the needs of the expanding community schools movement. This report, by focusing on the experience of the Redwood City School District, seeks to inform community school efforts in other parts of the country with insights into potential ways that community schools interact with students and families to improve student outcomes.

introduction and summary |


Background on community schools
Community schools 101
According to the Coalition for Community Schools,1 an alliance of national, state, and local organizations, community schools “purposefully integrate academic, health, and social services; youth and community development; and community engagement—drawing in school partners with resources to improve student and adult learning, strengthen families, and promote healthy communities.” Although not solely targeting academic outcomes, the theory of change behind community schools, summarized in the logic model developed by the Coalition for Community Schools in Figure 1, posits that providing wraparound supports for students’ social, physical, cognitive, and economic needs in the short term will aid schools in improving students’ academic outcomes in the long term. Their strategy has been touted as a means for closing the achievement gap by providing compensatory services for underserved students who do not otherwise have access to the services that community schools provide. Already community schools have been implemented on a large scale in several areas, including New York and Chicago. (Arne Duncan, former CEO of Chicago Public Schools, has advocated for the expansion of community schools nationwide in his current role as U.S. Secretary of Education.2) The Oakland, Calif., Unified School District is also in the process of implementing a district-wide community school initiative. At least 5,000 schools in the United States identify as community schools. Based on these numbers, it is evident that the community school strategy is fast becoming a popular way to meet the needs of students that extend beyond the classroom.3

Background on community schools |



How community schools work
Community schools logic model
Inputs • Support from schools and community • Sufficient staff (expertise + availability) • Sufficient resources (funding, facilities) • Available/relevant partners • Initiative level infrastructure • Community school coordinator What can happen at commuity schools? • Family engagement • Extended learning opportunities/youth development • Health, mental health, and social services; family support • Social and emotional learning • Ealry childhood development • Adult education • Professional development for teachers, school staff, and partners • Activities that increase linkages between schools and partners Outputs • Improved relations with and support for families • Comprehensive services appropriately targeted • Integrated academic enrichment and social services to support children’s intellectual, social, emotional, and physical development • High-quality, engaging, instructional programs • Partner integration into school day Short-term (proximal) • Children are ready to enter school • Students attend school consistently • School climate is safe, supportive, and inviting • Students are actively involved in learning and their community • Families are increasingly involved in their children’s education • Families are engaged in own development • Schools have sustaining relationships with partners • Families provide supportive living environment • Enhanced community well-being Long-term (distal) • Students succeed academically • Students are healthy physically, socially and emotionally • Families are socially and economically sound • Communities are desirable places to live Impact • Students graduate ready for college, careers, and citizenship

Source: Coalition for Community Schools

In addition, community schools share several common characteristics. First, they are built on the following five key conditions for learning, identified by the Coalition for Community Schools:4 • Rigorous core instructional program • Student motivation and engagement in learning • Provision of services to support students’ basic physical and mental needs • Mutual respect between families and school staff • Community engagement Community school leaders intentionally form strategic partnerships with public agencies, nonprofits, and private businesses in the community to create these conditions for learning. They leverage these partnerships to turn the school into a


center for american progress | positive student outcomes in community schools

hub of services, activities, and programs that create the necessary conditions for students to be successful. Community schools, however, take on many different configurations according to the needs of their particular population and the resources available to them, meaning that no two community schools—even in the same community—look alike. Community schools may offer a variety of services on the school campus, and these often are provided through partnerships with community-based organizations. Typically there is a community school coordinator who oversees the services. This person may be from a lead agency that partners with the school or a school district employee. And as a rule, schools generally transition slowly into becoming fullservice community schools because developing the set of partnerships that make up a community school takes time. Some community schools have evolved out of previous school reform strategies, such as school-based health centers, whereas others set out directly to become community schools. Recognizing these differences, the Children’s Aid Society, a charity that provides services to families in poverty, provides a useful framework for thinking about the stages of development of community schools, ranging from exploring to emerging to maturing to excelling.5

Current research on community schools
Community schools seem like an obvious approach to improving student outcomes, but analyzing their effects on student outcomes poses an inherent challenge. The robust network of integrated services and programs offered to both students and their families does not easily lend itself to a traditional evaluation research design that simply compares outcomes of participants to nonparticipants. As the Children’s Aid Society states, community schools are a “strategy, not a program.”6 In addition, community schools involve programs and services from a variety of service providers who may collect data differently (or not at all) and use different systems to store data. Moreover, schools may not have an infrastructure for sharing data with nondistrict partners, a reality that complicates gathering consistent and complete data. Also, the heterogeneity of community school implementation noted above makes it difficult to accurately examine outcomes across multiple community schools. Possibly due to these challenges, little rigorous research exists on community schools, with no studies appearing in peer-reviewed journals.7 The research that does exist tends to focus either very broadly on school-wide effects or very narrowly on a

Background on community schools |


specific program at a community school. There are several studies comparing average achievement scores or other outcomes for community schools to other schools.8 These descriptive studies, however, cannot explain how and for whom community schools affect outcomes, and they also attribute changes in outcomes to community schools when other contextual factors, such as neighborhood safety or student demographics, could also have had an influence. Further, descriptive studies fail to isolate the effects that may be due to community schools from these kinds of contextual factors that affect school operations as well as student and family participation. The other type of commonly cited research in the community school field focuses on evaluations of individual community school programs. There is strong research, for example, to support the efficacy of individual programmatic components, such as after-school programs,9 family engagement,10 and school-based health centers.11 The positive outcomes documented in these studies suggest that a wraparound approach such as community schools would benefit students and families. But without rigorous research focused on that coordinated approach, there is little information about the additive effects of programs, the importance of coordination, or the ways that they interact to improve short-term outcomes that may be linked to longer-term outcomes. There are several recent notable exceptions in the recent body of research on community schools. Communities in Schools (CIS), a national organization that provides wraparound services to students in nearly 200 schools nationwide, recently disseminated results from a methodologically rigorous, multi-year study that used both school-level analyses and student-level randomized controlled trials to evaluate community schools nationwide. This study found positive effects for CIS schools on dropout and retention, academic test scores, school attendance, discipline, and student attitudes, with the strongest effects for schools that had the highest-quality implementation of the CIS model.12 Another evaluation, of the Tulsa, Okla., community school initiative, linked the level of implementation of the community school strategy to students’ sense of collective trust at school and found that this sense of trust was linked to higher academic achievement scores.13 Seeking both to add rigorous research to the community schools field and to help local partners understand and strengthen their community schools, the John W. Gardner Center for Youth and Their Communities at Stanford University, or JGC, has been working with the community school initiative in Redwood City, Calif., for the last four years to document both the effects and context of community schools in this local initiative.


center for american progress | positive student outcomes in community schools

Redwood City context and research design
History of the Redwood City community schools
Redwood City is a diverse community south of San Francisco that has wide disparities in income and socioeconomic status, with some very affluent areas and others that are high-poverty and predominantly Latino and immigrant. The decision to create community schools in Redwood City grew out of a desire for greater equity for disadvantaged city youth and from a history of collaboration among city, county, and school district leaders. Nearly two decades ago, in 1995, representatives from each entity came together to support youth development and education through a collective body known as Redwood City 2020. Shortly after its inception, this collaborative secured funding from the California Healthy Start Initiative, a grant program aimed at funding schoolintegrated services to address students’ and families’ physical, social, emotional, and learning needs, and established four Family Resource Centers at the public elementary and middle schools serving the most disadvantaged Redwood City neighborhoods. The partners’ long-term commitment to strengthening and sustaining these four sites gave rise to the Academy for Community Schools Development, a multiyear initiative in partnership with the JGC to build community schools in the district. The academy provided an organizing framework for the four pilot schools as they began the process of transforming their campuses into effective and sustainable community schools. As part of the initiative, stakeholders from each school, including Family Resource Center staff, teachers, parents, and students, came together to develop and implement community school plans that emphasized high-quality academic supports, comprehensive youth and family resources, shared leadership between school administrators and community school coordinators, and youth engagement. Over time, the resource center staff shed their old titles and reemerged as community school coordinators, overseen by the director of school-community partnerships at the district level and working in direct partnership with their school principal.

redwood city context and research design |


Today the community schools initiative in Redwood City is owned by the school district and continues to receive guidance and resources from Redwood City 2020. The initiative has garnered wide support beyond its original proponents and is currently expanding from the original four schools to an additional two schools that are in different stages of community school development. The initiative now includes the four original Family Resource Center sites, one school in its second year of community school implementation, and one school that began implementation in the 2011-12 school year. The five existing community schools enrolled a total of 3,666 students in 2010-11. Students enrolled at these schools in 2010-11 were 89 percent Latino and 5 percent white, included 68 percent receiving subsidized meals under the National School Lunch Program, and included 67 percent who were English learners. In comparison, students at the rest of the Redwood City School District schools were 58 percent Latino and 31 percent white; 42 percent received subsidized meals and 38 percent were English learners.

Building the community school database
With the current literature on community schools in mind and a desire among partner agencies in Redwood City to improve understanding of their local community school initiative, the JGC, Redwood City 2020, and the city school district in 2007 jointly initiated a collaborative research process. The key questions that have guided the research throughout this process are: • How many and which students and parents access programs, and in what combinations, at the community schools? • What is the relationship between participation in community school services and student outcomes? The research uses the Youth Data Archive, a JGC initiative that links data on individual students collected by multiple agencies in order to answer questions that would not be possible to answer by any single agency alone. Agencies that are partners in the Youth Data Archive share individually identified data with the JGC and its researchers, who in turn link these data on individual youth across agencies to understand how, in the aggregate, activities in multiple settings relate to each other and to student outcomes and attitudes.14


center for american progress | positive student outcomes in community schools

For the purposes of this paper, we matched data from the following sources: • School district administrative data, including: – Student demographic characteristics. The analysis uses student demographic data from the administrative data collected by the Redwood City School District from the 2006-07 through 2010-11 academic years. Student demographic data available from each year include gender, ethnicity, English language proficiency status, parent education level, subsidized meal participation, and special or gifted education status. – School attendance. School attendance data provide the number of days students were present at school as well as the number of days absent each year. – Academic achievement. The measure of academic achievement for the analysis is standardized test scores, also provided from the district’s administrative data. The analysis uses two different tests—the California Standards Test (CST) for math and English language arts (ELA), which all students take each year from grades 2 to 8; and the California English Language Development Test (CELDT), which students who are not English proficient take starting in kindergarten until they are redesignated as English proficient. Because the CST and CELDT scores are not comparable across years or grades, all test scores are converted to percentiles, normed on the distribution of scores for all students in the state, by grade and year.

• Program attendance. The program attendance data is collected by both outside providers and school staff. The extent of participation data collected varied across programs. Support program data is maintained in the district administrative data system, and participation in these programs is captured only as a yes or no response. Extended learning and family engagement program data are maintained by program providers and reported to a centralized data system administered by the school district. This system contains daily attendance data for each program as well as student demographic data. Daily attendance data only became available for extended learning starting in the 2008-09 school year and for family engagement programs in 2010-11. Prior to those years, participation was tracked only as either a yes or a no.

redwood city context and research design |


• Survey data on student motivation and experiences at school. Finally, the analysis uses data from a survey developed and collected by JGC researchers as part of a larger survey on students’ motivation related to learning and their experiences of related practices in the classroom.15 The survey included all sixth- through eighth-grade students in the district and was administered in the spring of 2009, 2010, and 2011. The survey focused on three key measures: (1) students’ reported feelings of being supported at school, (2) students’ reported feelings of confidence in their academic abilities, and (3) their motivation to learn. Research shows that teacher practices that enforce mutual respect among students, such as showing respect for all opinions and not allowing students to put each other down, lead students to feel supported in their school environment.16 Supportive environments, in turn, promote motivation to learn when coupled with teachers’ practices that promote motivation and academic confidence, such as setting high but reasonable expectations for students, avoiding competition, and helping students to set goals.17 Academic motivation and confidence have a strong, positive influence on academic achievement.18 Thus, the data from the middle school motivation survey provide insights into short-term outcomes that link program participation at community schools to long-term academic outcomes.

Analytic approach
With nearly 250 programs and events at the Redwood City School District’s community schools in the 2010-11 school year, we classified programs according to three main strategy areas—family engagement, extended learning, and support— with program subcategories as shown in Figure 2. This categorization, which was devised with input from the community school coordinators, greatly enhanced our ability to conceptualize students’ level of involvement in community schools across the network of available programs and supports.


center for american progress | positive student outcomes in community schools


Redwood City school district programs
Community school strategies and programs categories
Strategy area
Family engagement

Program category
Parent leadership

• School site council • PTO/PTA • Leadership coaching • ESL classes • Computer classes • Volunteer activities • Outreach • Movie nights • Play dates • School socials • Report card night • Morning coffees with the principal • Back-to-school nights • After-school programs • Summer/intercession programs • Conflict managers • Legal education program • Individual Counseling • Group Counseling • Case Management • Bus Passes • Uniform Help • Holiday Gift Cards

Parent education Parent volunteer Parent events

School-home communication

Extended learning


Youth Leadership




After examining the extent of participation in community school programs, we next examined the links between program participation and student outcomes. Understanding the effects of community school programming is difficult because there are underlying factors about students and families that may influence both their decision to participate in programming as well as students’ academic outcomes. If parents, for example, have favorable attitudes toward school, they are probably more likely to participate in family engagement opportunities. The communication of these attitudes to their children likely promotes motivation in students that leads to working harder in school and higher achievement. Although it is impossible to claim that certain programs cause particular outcomes in this type of analysis, we took several steps to ensure that we were isolating the role of community schools from other factors.

redwood city context and research design |


First, we estimated effects using multi-level individual growth models, which measure average differences between students’ initial scores as well as their change over time, allowing us to differentiate between any preexisting differences in student outcomes. Instead of comparing outcomes for participants and nonparticipants in one year, these models estimate the difference in rate of growth of that outcome over multiple years, taking into account any differences in starting points between participants and nonparticipants. Second, knowing that there are differences among the five Redwood City community schools in terms of school climate, policies, and the neighborhoods that the schools serve, we included in our models school-level effects to take into account these outside effects and better isolate the link between community school services and student outcomes. Finally, by including student survey data, we controlled for preexisting student attitudes about school or learning that potentially underlie achievement and that often are not measured in other research.


center for american progress | positive student outcomes in community schools

Study findings
Participation in community school programs
Most of the students enrolled in a Redwood City community school in 201011 had some involvement with the supplemental programs provided by the community schools. Figure 3 shows that participation in the three main areas of community school programming—extended learning, family engagement, and support—has generally increased over time, which is consistent with enhanced implementation over time. The rate of overall participation in any program across all five schools fell from 72 percent to 70 percent, but this includes the addition of a community school in the emerging phase of program implementation in 2010-11. As is clearly shown, extended learning has consistently had the highest level of participation of all strategy areas. Participation in support programs, however, particularly in the four established community schools, has increased more than other categories. Programs at the community schools generally served the most socioeconomically disadvantaged students. This finding is important given the intended focus on community schools as a strategy for closing the achievement gap. As Figure 4 shows, students who accessed support services were significantly more likely to participate in the National School Lunch Program compared with participants in other programs and students with no program participation, and they were least likely to have a parent who completed college or scored proficient on the


Participation in community school supplemental programs
Percent of students enrolled in a community school participating in each strategy area, 2007-08 to 2010-11
80% 70% 60% 50% 40% 30% 20% 10% 0%

48 47 38 39





72 70 49 34 28 30


26 30 28 7

39 37 26

Extended Learning

Family Engagement


Any Participation

No Participation

Source: Sebastian Castrechini and Rebecca London, John W. Gardner Center for Youth and Their Communities at Stanford University

study findings |


California Standards Test. The Redwood City community schools served high rates of English learner students, and family engagement programs in particular had significantly higher participation of parents of students who were English learners compared with participants in other programs and students with no program participation.

Community school student breakdown
Student background characteristics by community school program participation in 2010-11
Extended Learning
50.3% 49.7% 1.3% 89.6% 5.7% 3.4% 11.7% 3.2% 65.1% 69.1% 31.7% 48.4% 6.1% 49.2% 34.3% 1,728

  Female Male African American Latino White Other ethnicity Special Education Gifted and Talented English learner Free/reduced lunch Parents HS diploma Parents no HS Parents college Proficient 2009-10 Math Proficient 2009-10 ELA Number of Students

Family Engagement
47.7% 52.3% 0.3% 95.2% 2.8% 1.7% 9.8% 2.2% 75.4% 58.4% 24.8% 52.1% 4.7% 56.4% 36.1% 1,019

48.7% 51.3% 1.6% 92.2% 3.2% 3.0% 13.3% 1.1% 72.3% 71.3% 26.7% 50.4% 2.9% 44.2% 28.5% 1,341

Any Program Participation
49.2% 50.8% 1.3% 90.5% 4.7% 3.5% 12.0% 2.6% 67.8% 68.0% 29.4% 48.3% 5.0% 50.1% 33.7% 2,578

No Participation
44.2% 55.8% 1.7% 84.6% 6.4% 7.4% 11.4% 2.3% 64.8% 65.3% 27.8% 40.3% 6.4% 45.4% 31.9% 1,088

Source: Sebastian Castrechini and Rebecca London, John W. Gardner Center for Youth and Their Communities at Stanford University

Across all schools, many students accessed programs from more than one strategy area. This finding speaks to the wraparound nature of services at community schools and the importance of looking at outcomes from a multi-agency perspective. As Figure 5 shows, approximately 8 percent of students enrolled in a community school accessed programs from all three strategy areas in 2010-11, 24 percent accessed two services in different combinations, and 38 percent accessed just one service. Among those accessing two services, family engagement participants were significantly more likely than students who participated in other programs


center for american progress | positive student outcomes in community schools

to participate in at least one other strategy area, with only 6 percent of students accessing family engagement alone compared with 19 percent and 12 percent for extended learning and support, respectively. Extended learning and support were the combination of programs most frequently accessed together.

Program participation rates in single strategy areas and combinations of strategy areas in 2010-11
  Accessed One Strategy Area Only Extended Learning Only Family Engagement Only Support Only Accessed Two Strategy Areas Extended Learning and Family Engagement Extended Learning and Support Family Engagement and Support Accessed All Three Strategy Areas No Participation Number of Students

Participation Rate
37.6% 19.1% 6.2% 12.3% 24.4% 8.5% 11.2% 4.7% 8.4% 29.7% 3,666

Source: Sebastian Castrechini and Rebecca London, John W. Gardner Center for Youth and Their Communities at Stanford University

Linking community school participation to student achievement
We examined the extent to which participation in community school programs was linked to student achievement outcomes, after controlling for individual characteristics and school-level differences. As previously noted, there are many underlying factors about which we do not have data that may affect both student participation in programs as well as student outcomes. Therefore, the findings in this section show relationships between participation and outcomes, independent of student background characteristics, but should not be taken to imply a causal relationship.

Community school participation and English language development
In terms of academics, we found the strongest links between family engagement and gains in English Language Development scores for English learner students.

study findings |


Figure 6 shows differences in California English Language Development Test percentiles based on years of family engagement over the five years from 2006-07 to 2010-11. Among English learners who scored at levels 1 or 2 (beginning and early intermediate) on the CELDT in 2006-07, students with family engagement in three or more of the next four years gained, on average, 3.5 points per year more than students with no family engagement, which was a statistically significant difference resulting in a 12-point gap by 2010-11. Students with one or two years of family engagement showed smaller gains. Our findings linking family engagement and achievement are consistent with other research that has shown similar results19 and are important in that they provide evidence for using the community school strategy as a means of improving achievement for a population of students that is growing and has traditionally had low achievement. The relationship between participation in community school programs Redwood City community school gains and English language development in English language proficiency scores differs over time for CELDT percentile trajectories for English learner students at CELDT levels students. In the early grades, there 1 or 2 in 2006-07, by years of family engagement for students enrolled is a strong relationship between at community schools family engagement and California 60 55 English Language Development 0 years 3 or 4 years 50 Test scores. Among middle school 46 43 students, however, it does not appear 40 37 37 that family engagement alone is 27 30 31 sufficient to support English language 20 25 20 progression. This is likely because 18 family engagement levels drop off 10 substantially in the Redwood City 0 School District’s middle schools, as is 2006-07 2007-08 2008-09 2009-10 2010-11 typical for middle schools nationally, Source: Sebastian Castrechini and Rebecca London, making it difficult to examine a link John W. Gardner Center for Youth and Their Communities at Stanford University between family engagement and student outcomes. In addition, many students have a hard time progressing past CELDT level 3 (intermediate), which requires the acquisition of academic English,20 and it becomes harder to become proficient as the content and academic language demands increase in middle school. Continuing gains in California English Language Development Test scores in middle school were instead associated with frequent participation in extended learning programs.


center for american progress | positive student outcomes in community schools

Among students from this same cohort of students at CELDT levels 1 and 2 in 2006-07 who were still at level 3 or lower as middle school students in 2009-10, participating in extended learning programs for at least 120 days was associated with a 10 percent increased likelihood of reaching level 4 (early advanced) in 2010-11, a statistically significant difference.

Community school participation and math and English Language Arts achievement
There were no direct, statistically significant links between community school program participation and the California Standards Test scores in either math or English Language Arts. Students’ motivation to learn and academic confidence were critical to their math and English Language Arts achievement. The achievement effects related to community school program participation, particularly in extended learning, only existed for students who developed positive attitudes, which were associated with participation in extended learning opportunities. As it turns out, these attitudes were strong predictors of California Standards Test math achievement for everyone but had particularly strong effects for English learner students. Middle school students who reported confidence in their academic abilities gained, on average, approximately 7 points in math and 6 points in English Language Arts scores compared with students who did not have positive reports of their confidence; this takes into account student background characteristics as well as prior achievement. It is important to note, however, that students with high levels of extended learning participation also entered the year with significantly higher senses of confidence in their abilities. Unlike other community school strategy areas, the link between support programs and student outcomes is not always the same in this analysis. The array of support programs and the extent of the population that they have reached have increased over time, so what it means to have received support services changes over time. Also because these services target the most disadvantaged students and families, students receiving services may have life circumstances that are beyond what we can accurately control for with available data on subsidized school lunch participation or parent education levels. This makes it difficult to reliably compare the students who received support services with those who did not.

study findings |


The role of community school participation in motivating learning
Given the important role of student academic motivation and confidence in explaining links between community school program participation and achievement, we examined the relationship between program participation and these attitudes more deeply. As mentioned above, research shows that creating supportive environments at school—where students feel that they belong and that other students and adults are there to help them—is critical to developing academic motivation and confidence, both of which have been shown to promote academic achievement. Examining raw data on the survey responses over three years for all community school students who entered middle school in 2008-09 shows that students’ feelings of support declined over their time in middle school (see Figure 7). The first set of bars in the figure shows that reports of feelFIGURE 7 ing supported among all students in this cohort decreased from Students who view their schools 35 percent in sixth- grade to 31 percent in seventh grade and 28 as supportive percent in eighth grade. English learners in this same cohort (the Percent of students reporting feeling second set of bars) had initially higher reports of feeling supported cared for in the community schools, in sixth grade compared with other sixth-graders in the community 2008-09 to 2010-11 schools, but the difference narrowed over time, with English learn100 90 ers’ reports of their schools being supportive declining over the 6th grade 7th grade 8th grade 80 next two years more, on average, than other students. 70 Again linking community school program participation data to survey data, we found that community school programs appear to play a role in establishing and maintaining students’ feelings of being supported at school, potentially buffering against the declines shown in All students in EL students in Figure 7. There are several reasons for this. First, students whose par2008-09 6th grade 2008-09 6th grade ents were engaged in elementary school reported significantly higher Source: Sebastian Castrechini and Rebecca London, initial ratings of their schools’ supportiveness in sixth grade comJohn W. Gardner Center for Youth and Their Communities at Stanford University pared with students with no family engagement program participation in elementary school. Students with one or more years of family engagement in elementary school were 19 percent more likely to report feeling supported in sixth grade.
35% 31% 28% 37% 30% 60 50 40 30 20 10 0 45%

Second, extended learning participation during middle school, rather than family engagement, was linked to significant increases in students’ ratings of their schools’ supportiveness over time. Students with at least 120 days of participation were 10 percent more likely to report that their schools are supportive environments compared with students with no extended learning participation. These findings


center for american progress | positive student outcomes in community schools

parallel the programmatic areas that showed links to gains in California English Language Development Test scores in the previous section. Not surprisingly, consistent with other research,21 we found that students’ feeling supported had a significant effect on their motivation and confidence. All of these findings are statistically significant and take into account other student and school background characteristics, students’ previous survey results, and their reports of experiences in their classrooms. These findings, however, do not necessarily mean that community school programming is the cause of differences in survey results because there could be other underlying factors for which data were not available, such as parents’ attitudes toward school, that influence both their engagement and their children’s attitudes toward school.

study findings |


Implications for policy, practice, and future research
The study detailed in this report used linked data on community school program participation, student achievement, and surveys of students’ attitudes and beliefs about learning, to examine community schools and student outcomes. We found that participation in extended learning and family engagement opportunities over time is associated with achievement gains. We also found evidence that students’ own motivation to learn is associated with participation in community school programs, which suggests a possible mediating effect. These results point to some ways that improving data and research on community schools can further advance the field by informing policy and practice. Importantly, conducting high-quality community schools research may require amending or creating data systems so that they capture student and family participation and can link to students’ academic outcomes. Using data to inform community schools practice is a critical component of our work and has proven to be a valuable practice for the Redwood City School District. And while the findings in this paper are specific to the Redwood City community schools included in the study, they do suggest that the community school strategy has promise. Realizing this promise, however, will require action on the part of policymakers as well as advances in data analysis related to community schools. We outline below some specific policy recommendations for broadening the reach of community schools, leveraging community schools to improve outcomes for English learners, strengthening students’ motivation at school, and improving research and practice on community schools through data sharing.

Implications for federal and state policy
At the federal and state levels, policymakers must continue to advocate for the expansion of community schools, even in the current climate of shrinking budgets and tight resources, to realize the potential that community schools have in

implications for policy, practice, and future research |


influencing positive academic outcomes for students. This will require a shift in the common perception that noninstructional services provided at schools are unnecessary add-ons and instead see them as essential contributors to students’ learning. There is recent evidence that teachers and school administrators see community schools services as promoting learning by facilitating better connections with families, removing factors in students’ lives that are frequently barriers to learning, decreasing student mobility, and reducing teacher stress and burnout.22 Also, community schools can and must be seen as an efficient use of resources. They provide a hub for coordinating services that often already exist, which creates an opportunity for integration and collaboration among service providers, reducing redundancy, and making optimal use of the resources available. In fact, a recent report by the Coalition for Community Schools found that every $1 invested by school districts on community schools leverages $3 in additional resources or funding from other stakeholders.23 Another key implication, given that our findings show the importance of Englishlearner students accessing multiple services at community schools, is to offer inducements that encourage multi-agency collaboration to integrate services at community schools. Aligning curricula between after-school programs and classrooms, creating cross-program communication structures for providers to make referrals and share data, and integrating student programs with family engagement opportunities can all promote greater efficiency and effectiveness of service delivery. This may be especially the case for immigrant or non-English speaking parents who often are less comfortable or less able to seek out services elsewhere. With the numbers of English learners in U.S. schools steadily increasing and many policy efforts focused on improving achievement for these students, the findings from this study are important because they suggest that community schools could be another strategy to reach this goal. The recent growth of community schools and the support that they have received from federal and state policymakers have led to the creation of several funding sources, such as the Full Service Community School grants, Promise Neighborhoods, and 21st Century Learning Centers.24 Federal and state policy can continue to help by streamlining funding sources that go toward community schools in order to prevent fractured service delivery and competition among service providers within a community. The results of this study show that the English learners stand to benefit from the combination of both family engagement and extended learning services, meaning that strengthening the ties between these services could have long-term academic benefits for students.


center for american progress | positive student outcomes in community schools

Beyond government support for community schools, the philanthropic community has a role in fostering collaborative structures instead of competition for resources and participants. Foundations, for example, could require joint reporting instead of individual program evaluations that would provide an incentive for programs to collaborate and share information. Further, fractured funding may have a detrimental effect on collaboration and data collection on shared goals. As with collaborative service delivery, collaborative data collection and analysis will prove more efficient and provide richer results than providers working independently. Evidence in this study and others shows that when students feel connected to their schools, there are important positive implications for their academic motivation and confidence, both of which have strong links to academic performance. Promoting student social and emotional development has not been a major policy priority at the national level. Findings showing a strong link between positive student attitudes and supplemental community school programs suggest that community schools, through the multiple opportunities they provide for positive interactions between students and their schools, could be a promising strategy for fostering positive attitudes that promote achievement. This means, however, that it is important for the many staff that students and families interact with in community schools—both in the classroom and in outside programs—to consistently and uniformly implement the practices known to foster positive attitudes. Fortunately, there are a number of opportunities for federal and state policymakers to establish policies to encourage the support of positive student attitudes. Illinois, for instance, developed a plan to include social and emotional development standards as part of the state’s learning standards with key goals focused on self-management, positive relationships, and personal responsibility.25 This serves as just one example of how nonacademic instructional practice can be incorporated into state standards and accountability systems. Additionally, this study shows the importance of data and research in improving community schools. Federal and state policymakers can enable and facilitate the type of cross-agency analysis presented in this report in several ways. One way is by simplifying and easing regulations on data sharing to allow the multiple providers at community schools to link and collectively analyze data on the students they serve in common. Although current laws do allow for this type of data sharing, the regulations in the Family Educational Rights and Privacy Act (FERPA) and the Health Insurance Portability and Accountability Act (HIPAA) are often cited as reasons for agencies to not share data with partner agencies for fear of repercussions.

implications for policy, practice, and future research |


The U.S. Department of Education recently released new regulations making it easier to use educational data, but fear of violating HIPAA’s data confidentiality requirements still remains a barrier for using health or social service data. Further state law often layers on additional regulations and procedures. Easing regulations does not mean that data confidentiality for students and families should be compromised, but ways need to be found to make data-sharing less complex and daunting. Federal and state funding for data infrastructure is essential to making this a reality.

Implications for districts and schools
Policies that support community school efforts at the district level, instead of isolated community schools within larger communities, will maximize the potential that community schools. This is particularly important given that students in many cities and towns do not attend their neighborhood schools, and they may feed into schools in neighboring communities. The Coalition for Community Schools argues that scaling up community school initiatives to a system level ensures the greatest possible benefits for youth and their communities.26 The Oakland Unified School District recently adopted this approach, showing that a large school system can become a community school district. Superintendent Tony Smith’s strategy of gaining approval to turn Oakland into a “full-service community school district” focused on the importance of a system-level initiative for increasing equity among Oakland schools as a way of improving student outcomes, with the district playing a key role in facilitating school-level implementation. The implementation process included an intensive input process from over 350 community meetings as well as youth surveys and family focus groups. After this initial engagement process, the district set up a robust infrastructure of school and district committees to support the work and adopted a balanced scorecard system for measuring ongoing progress and providing accountability toward the initiatives’ goals.27 Smith attributes approval of his plan to selling it as a strategy to improve achievement and to the development of a concrete plan for implementation.28 Within community schools, family engagement is a strategy that requires farreaching collaboration to be successful. The findings from this report suggest that forging partnerships with families at community schools can have strong benefits for English learner students. This aligns with the broader family engagement


center for american progress | positive student outcomes in community schools

literature that shows that family engagement is linked to positive student outcomes.29 Schools and districts, however, face many challenges in engaging parents in meaningful ways. Maintaining engagement after students move on to middle school, for instance, has been a challenge both in Redwood City as well as in other communities nationwide.30 Still, community schools have an advantage because they can engage families through many access points where staff can refer them to additional opportunities and broaden engagement. And while the findings from this study show specific benefits for English learner students, parents with limited English proficiency frequently are reluctant to engage in school activities. Engagement strategies such as utilizing peer-to-peer parent mentoring and providing opportunities for parent input in decision making can help to improve parental involvement at school.31 Moreover, family engagement needs to extend beyond parents coming to school for programs or events.32 Family engagement both at school and at home is important to student success, and programs at community schools can provide a venue for parents to gain the skills and confidence to be more actively involved in their children’s education at home. The research in this report also supports the important role community schools can play in fostering supportive settings that promote student motivation. School and district leaders can capitalize on the potential achievement gains related to supportive school environments and motivation highlighted by this research in several ways. District superintendents or school boards, for instance, can take an explicit stance on the importance of practices that create supportive environments and promote student motivation by drafting district policies mandating school leaders to include these components in school improvement plans. Second, as research underscores the importance of teacher practices in fostering student motivation, district leaders can engage teachers in any planning efforts to institutionalize these practices. This is especially the case for community schools, where teachers are critical when it comes to providing input about specific classroom practices that align with the community school goals. District leaders can then ensure that staff members are using these practices by adapting current classroom or program observation rubrics used in teacher and program evaluations to hold staff accountable for using practices that promote student motivation. Such measures should be accompanied by ongoing professional development that includes teacher input in order to ensure student motivation improvement goals are part of school-wide goals for learning and achievement.33

implications for policy, practice, and future research |


Additionally, regardless of where district or school leaders are on the community school development continuum, they can invest in data collection efforts to better understand and track school climate and student motivation. Developing a system of indicators that includes a variety of measures of student success, including those focused specifically on academics and those focused on other aspects of positive youth development34, is critical for understanding the efficacy of community schools. Adopting a system of indicators will enable districts to reliably flag and intervene with students in need of support.35 Such measures are not typically included in school administrative records and would therefore require new data collection or a link to an existing one, for example, the California Healthy Kids Survey collected by the WestEd research group. Whether community schools are implemented at the district or school level, involved leaders play a critical role in building shared data systems to support community schools. Developing common goals and indicators that cut across programs can help to solidify partnerships. Redwood City 2020 is working toward this by developing common indicators for which programs are expected to collect and report, thereby downplaying individual program evaluations. Understanding how to target interventions to the appropriate students and parents is a particular challenge, as data systems are often designed more for accountability reporting than identifying student needs. This report clearly shows that program participants are those who came into the school year already feeling more strongly motivated and connected at school. The data also show that traditional demographic factors that are readily available in school data systems do not seem to fully describe the profile of students who are disconnected from their schools. These findings underscore the importance of prioritizing the needs of students who have not developed these positive attitudes. Sharing program participation records among providers can help practitioners at community schools to identify youth and parents who have a particular need for outreach. Additionally, sharing survey data on student attitudes and practices at school can provide important information about how educators and service providers can improve programs to better suit students’ needs and make them more appealing to their intended audiences. Enhancing recruitment strategies after identifying students in need of intervention is also crucial, particularly for older students. Research has shown that direct outreach and adapting programs to meet student needs and schedules becomes increasingly important as students progress beyond elementary school.36


center for american progress | positive student outcomes in community schools

Implications for community school researchers
Researchers play a key role in helping policymakers and practitioners understand, improve, and sustain the work of community schools, and it will be important to improve the quality of data collection and analysis to meet the needs of this advancing field. As previously noted, individual program evaluations, on the one hand, lose the richness of the wraparound service model that makes community schools unique, and simply looking at whole-school outcomes over time on the other hand masks potential differences within school and fails to explain how and for which students the community school strategy works. Although challenging, it is imperative that researchers develop improved structures that allow for: • Consistent, accurate data collection • Linking data and sharing it across community school initiatives • Analyzing strategies that mirror the complex ways in which community schools work while also communicating those data in a manner that is accessible to and actionable for both practitioners and policymakers Creating an infrastructure for sharing data and examining that data as a group that includes district and school personnel, service providers, and even community stakeholders is vital to the success of a school-wide analysis approach. Policymakers play a key role in encouraging collaborative data analysis as well as promoting policies that allow for data sharing. But they are not alone; researchers working with community schools data also have to play a role in building these collaborative partnerships. For one, capturing participation in all of the programs and events that happen at community schools is complicated and takes concerted efforts on the part of practitioners along with the expertise of researchers to help design relevant systems. Researchers also need to work closely with practitioners to balance the need for thorough and complete data on one hand, with the risk of a burdensome data collection process for practitioners on the other. Researchers and practitioners can make this happen by creating processes that meet the needs of both groups. It is important for researchers to engage stakeholders frequently through the research process, including after analysis is completed, to discuss results and implications. This step helps to build interest in and understanding of the findings as a means of improving practice as opposed to an imposed evaluation process. This approach also creates buy-in that ultimately helps to improve data collection and sharpen the research strategy. Additionally, researchers will need to develop

implications for policy, practice, and future research |


consistent ways of collecting and recording data to ensure that the data can later be used to look at participation and outcomes across the many program settings in which students and families participate at community schools. Once robust cross-agency data systems are established, the challenge remains to analyze the data in a way that is true to the community school strategy. The findings presented in this analysis provide a framework for examining the cumulative effects of the combination of multiple services and programs offered at community schools over time. This entails using a youth development framework for the analysis that looks at the combination of youth participation settings and analyzing multiple indicators to understand the progression of short-term to long-term outcomes. Our analysis of Redwood City’s community schools seeks to do this, and we believe it provides richer insights into the effects of community schools on student outcomes than isolated evaluations of the individual programs that comprise community schools. Still, challenges abound when searching for approaches that accurately mirror the complex ways in which services and programs at community schools interact. Moreover, each community school initiative is unique and will require approaches that are tailored to the realities on the ground. Some successful strategies we’ve identified based on our partnership in Redwood City include: • Involve stakeholders throughout the research process to ensure that the data analysis is framed in ways that match how they view community schools. • Use a logic model to guide the research so that it builds on an existing shared understanding of the goals of the community school. If there is not an existing logic model, develop one in partnership with the school or district to help to build understanding and buy-in for the research process. • Look at an array of indicators—beyond the traditionally used measures of achievement, graduation, and attendance data—to understand processes through which community schools improve long-term student outcomes (see Appendix 1 on indicators as well as the Community Schools Evaluation Toolkit, available at schools_evaluation_toolkit.aspx).


center for american progress | positive student outcomes in community schools

• Gather and have the flexibility to respond to feedback from district and school administrators, school personnel, and community stakeholders throughout the research process to ensure that the research design reflects practitioners’ impressions and that the findings are actionable. • Successful community engagement in research involves finding accessible ways to communicate complex analyses, a practice that is difficult but crucial to creating an environment in which community schools feel comfortable and able to use evidence to implement strategies and improve practice.

implications for policy, practice, and future research |


We believe the findings from this study contribute to the community school field by providing evidence of some important ways in which community schools link to student outcomes. The patterns of participation across the dense network of services and the relationships between programs and outcomes are complex, with some only materializing after several years of engaging in community school programming. Furthermore, the associations demonstrated between community schools and student outcomes do not necessarily mean that community schools give rise to these outcomes. Still, the data show some clear patterns linking participation to improvements in English language development for English learners. Underlying this connection, we found that community school participation is linked to students’ feeling that their schools support them as well as their confidence and motivation, which in turn link to longer-term improvements in achievement, particularly for English learners. Our findings are aligned with the community school logic model in that participation influences students’ short-term beliefs, which in turn affects their achievement. The research presented in this report influenced policy and practice at the community schools in Redwood City in several ways. First, examining program participant characteristics has spurred improvements in data collection by community school coordinators to ensure that the activities in which students and families participate are accurately captured. The careful examination of profiles of students and families participating in services has also stimulated discussions of strategies for targeting groups of nonparticipants. Additionally, the strong findings related to family engagement created the impetus for district-wide professional development workshops for community school coordinators and principals on family engagement practice. These have also prompted a follow-up qualitative study by the John W. Gardner Center for Youth and Their Communities on family engagement in the community schools to further understand successful practices for engaging families to improve

conclusion |


outcomes for youth. Although these findings and actions are specific to the Redwood City School District, the process by which Redwood City has utilized partnerships to collect data on their community schools and continue to use those data to inform policy and practice serve as a prime example for how other community school efforts can further their work to improve outcomes and equity for youth and communities.


center for american progress | positive student outcomes in community schools

Community school results framework mapped to indicators and potential data sources
Conditions for Learning
Early childhood development is fostered through high-quality, comprehensive programs that nurture learning and development.


Immunization rates Blood lead levels

Data Sources
School immunization records Health center records Health center records Parent surveys ECE center attendance records Kindergarten readiness tests Attendance records from parent programs School district achievement data School district achievement data After-school program/CBO records School district achievement data community college records School district achievement data School district human resources records School district attendance data School district discipline data School district attendance data Police/probation records Youth development surveys School counseling records After-school program participation Program participation records from CBOs Teacher grade books Student surveys Student surveys School physical fitness test scores Health center records Public health department records School nutrition audits Community eating environment assessments

Children are ready to enter school

Parents read to children Children attend early childhood programs Receptive vocabulary level Families connected to support networks/ services Standardized test scores Students demonstrate competencies via multiple modes

The school has a core instructional program with qualified teachers, a challenging curriculum, and high standards and expectations for students.

Students succeed academically

Graduation rates Dropout rates Teacher attendance rates Attendance rates Suspension rates

Students are actively involved in learning and the community Students are motivated and engaged in learning—both in school and in community settings, during and after school. Students are healthy physically, socially, and emotionally

Truancy rates Students reporting feeling connected to the school Percent of students engaged in and contributing to community Homework completion rates Percent of students demonstrating social and personal competencies Percent of students demonstrating well-being on a range of health indices (immunizations, obesity, vision, hearing, asthma, STDs, pregnancy, substance abuse) Students have access to good nutrition

appendix |


Community school results framework mapped to indicators and potential data sources
Conditions for Learning Results Indicators
Percent of families whose basic needs are met The basic physical, social, emotional, and economic needs of young people and their families are met. Students live and learn in stable and supportive environments Student mobility rates Percent of students reporting relationship with caring adults Incidence of bullying Incidence of school vandalism Percent of families who report involvement with children’s education There is mutual respect and effective collaboration among parents, families, and school staff. Families are actively involved in children’s education Percent of families who participate in parent teacher conferences Percent of families who report positive interactions with school Percent of teachers who report positive interaction with families The community is engaged in the school and promotes a school climate that is safe, supportive, and respectful and that connects students to a broader learning community. Employment/employability rates Rate of participation in adult education programs Rate of participation at school events Percent of residents with health insurance Neighborhood crime rates

Data Sources
Human Services Agency/DCFS records School attendance data Youth development program surveys Discipline records Discipline records Parent survey Parent program attendance records Parent surveys Teacher surveys Labor/economic statistics Program participation records Program participation records Public health department records Police department data

Communities are desirable places to live


center for american progress | positive student outcomes in community schools

About the authors
Sebastian Castrechini, EdM, is a Senior Policy Analyst at the John W. Gardner

Center for Youth and Their Communities at Stanford University, working primarily with the Youth Data Archive. Castrechini has been at the JGC for nearly four years and previously worked as both a school teacher and administrator. Castrechini’s areas of research expertise include community schools and youth fitness and wellness. He has published quantitative and GIS-based research in the Journal of School Health and Penn GSE Perspectives on Urban Education and presented at the National Institutes of Health, American Educational Research Association, and Coalition for Community Schools conferences. Castrechini holds a B.S. in secondary education in social studies from Penn State University and a Master’s of Education in education policy from the Harvard Graduate School of Education.
Rebecca A. London, Ph.D., is Senior Researcher at the John W. Gardner Center for Youth and Their Communities and the researcher overseeing all analyses conducted with the Youth Data Archive. Throughout her career, London’s research has bridged academia and policy, focusing on the policies and programs intended to serve low-income or disadvantaged families and youth. Using both qualitative and quantitative methods, London has conducted research on a variety of policy-relevant topics such as physical fitness and academic achievement, secondary to postsecondary transitions, the effects of after-school program participation, the digital divide for youth, the effects of welfare reform, college attendance among low-income mothers, children’s living arrangements, and health insurance for low-wage workers. Her work has been published in journals such as the Journal of Policy Analysis and Management, Social Science Quarterly, the Journal of Higher Education, Youth & Society, and Economic Journal. London holds a Ph.D. in Human Development and Social Policy from Northwestern University’s School of Education and Social Policy, an M.A. in Economics, also from Northwestern, and a B.A. in Economics from the University of Michigan. The John W. Gardner Center for Youth and Their Communities (JGC) at Stanford University partners with communities to develop leadership, conduct research,

and effect change to improve the lives of youth. Founded in 2001 by Dr. Milbrey McLaughlin, the JGC works both in local communities in the San Francisco Bay Area and nationally to promote community youth development.

about the authors |


We would like to thank Pat Brown at Redwood City 2020 and Sandra Portasio and Jan Christensen at the Redwood City School District for working with us to frame the analysis and interpret the results. Kara Dukakis, Amy Gerstein, and Milbrey McLaughlin provided valuable input in earlier stages of the work. We thank participants at the American Education Research Association 2011 Annual Conference for comments on an earlier version of this work. We also thank the Center on American Progress for commissioning this paper and Redwood City 2020, the Redwood City School District, the Walton Family Foundation, and William V. Power and Ellen W. Power Family Trust for funding our research in Redwood City. The Center for American Progress thanks the JPMorgan Chase Foundation for generously providing support for this paper.


center for american progress | positive student outcomes in community schools

1 coalition for community schools, “community schools research Brief 09” (2009), available at assets/1/assetmanager/ccs%20research%20report2009.pdf. shital c. shah and others, “community schools evaluation toolkit: a starter guide for community school staff who want to use data to tell their story and improve their community schools” (Washington: institute for educational Leadership, 2009). coalition for community schools, “frequently asked Questions about community schools,” available at martin J. Blank and others, “making the difference: research and practice in community schools” (Washington: coalition for community schools, 2003). Janice chu-Zhu, “national and international adaptations of the children’s aid society model,” in J.g. dryfoos, J. Quinn, and c. Barkin, eds., Community schools in Action: Lessons from a Decade of Practice (new York: oxford University press, 2005). Jane Quinn, “community schools: a strategy, not a problem” (new York: the children’s aid society), available at assets/docs/he/communityschoolspaper.pdf. curt m. adams, “the community school effect: evidence from an evaluation of the tulsa area community school initiative” (tulsa: oklahoma center for education policy, 2010); reuben s. Jacobson and shital c. shah, “the community schools evaluation toolkit: moving the research agenda forward” (philadelphia: coalition for community schools national forum, 2010). coalition for community schools, “community schools research brief 09”; martin J. Blank and others, “making the difference: research and practice in community schools”; coalition for community schools, “community schools—results that turn around failing schools” (2010). Kathleen m. morrissey and ronald J. Werner-Wilson, “the relationship between out-of-school activities and positive Youth development: an investigation of the influences of communities and family,” Adolescence 40 (157) (2005):67-85; deborah L. Vandell, elizabeth r. reisner, and Kim m. pierce, “outcomes Linked to high-Quality afterschool programs: Longitudinal findings from the study of promising afterschool programs” (Washington: policy studies associates, 2007). 16 helen patrick, allison m. ryan, and avi Kaplan, “early adolescents’ perceptions of the classroom social environment, motivational Beliefs, and engagement,” Journal of Educational Psychology 99(1) (2007): 83-98. 17 Barbara gross davis, Tools for Teaching (san francisco: Jossey-Bass, 1993). 18 ming-te Wang and Jacquelynne s. eccles, “adolescent Behavioral, emotional, and cognitive engagement trajectories in school and their differential relations to educational success,” Journal of Research on Adolescence (2011). 19 Joyce L. epstein and steven B. sheldon, “moving forward: ideas for research on school, family, and community partnership,” in c.f. conrad and r. serlin, eds., SAGE Handbook for Research in Education: Engaging Ideas and Enriching Inquiry (thousand oaks: sage publications, 2006). 20 Kenji hakuta, Yuko goto Butler, and daria Witt, “how Long does it take english Learners to attain proficiency?” (davis, calif.: University of california Linguistic minority research institute, 2000). 21 ming-te Wang and Jacquelynne s. eccles, “adolescent Behavioral, emotional, and cognitive engagement trajectories in school and their differential relations to educational success” Journal of Research on Adolescence (2011). 22 theodora chang and calyssa Lawyer, “Lightening the Load: a Look at four Ways that community schools can support effective teaching”( Washington: center for american progress, 2012). 23 martin J. Blank and others, “financing community schools: Leveraging resources to support student success” (Washington: coalition for community schools, 2010). 24 isabel owen, “Breaking the mold: combining community schools wtih expanded Learning time to help educationally disadvantaged students” (Washington: center for american progress, 2010). 25 illinois state Board of education, “illinois Learning standards: social/ emotional Learning (seL),” available at 26 atelia melaville, reuben Jacobson, and martin J. Blank, “scaling up school and community partnerships: the community schools strategy” (Washington: coalition for community schools, 2011). 27 oakland Unified school district, “community schools, thriving students: a five Year strategic plan” (2011). 28 coalition for community schools, “creating a full-service community school district: interview with tony smith, superintendent of oakland Unified school district” (2011), available at http://www. interview%20template1.pdf#xml= %3a%5cdtsearch%5ccommunityschools&hitcount=7&hits=20+21 +22+23+24+25+26+&hc=21&req=%22creating+a+full%2dservice +community+school+district. 29 anne t. henderson and Karen L. mapp, “a new Wave of evidence: the impact of school, family, and community connections on student achievement” (austin: national center for family and community connections with schools, 2002); Joyce L. epstein and steven B. sheldon, “moving forward: ideas for research on school, family, and pommunity partnership,” in c.f. conrad and r. serlin, eds.,. SAGE Handbook for Research in Education: Engaging Ideas and Enriching Inquiry (thousand oaks: sage publications, 2006). 30 Jaana Juvonen and others, “focus on the Wonder Years: challenges facing the american middle school (santa monica: rand education, 2004).









10 anne t. henderson and Karen L. mapp, “a new Wave of evidence: the impact of school, family, and community connections on student achievement” (austin: national center for family and community connections with schools, 2002). 11 sarah c. Walker and others, “impact of school-Based health center Use on academic outcomes,” Journal of Adolescent Health 46(3) (2010):251-257; s.p. geierstanger and others, “school-based health centers and academic performance: research, challenges, and recommendations,” Journal of School Health 74(9) (2004):347-352. 12 icf international, “communities in schools national evaluation: five Year summary report” (2010). 13 curt m.adams, “the community school effect: evidence from an evaluation of the tulsa area community school initiative” (tulsa: oklahoma center for education policy, 2010). 14 more information on the Youth data archive may be found at http:// 15 Karen strobel, “practices that promote middle school students’ motivation and achievement” (palo alto, calif.: the John W. gardner center for Youth and their communities at stanford University, 2010).

endnotes |


31 Karen L. mapp and soo hong, “debunking the myth of the hardto-reach parent,” in sandra L. christenson and amy L. reschly, eds., Handbook of School-Family Partnerships, (new York: routledge, 2009). 32 organization for economic cooperation and development,” What parents can do to help their children succeed” (2011). 33 richard f. elmore and deanna Burney, “investing in teacher Learning: staff development and instructional improvement in community school district #2, new York city” (new York: national commission on teaching and america’s future, 1997).

34 Kara dukakis and others, “positive Youth development: individual, setting, and system Level indicators” (palo alto, calif.: John W. gardner center for Youth and their communities at stanford University, 2009). 35 John W. gardner center for Youth and their communities, “early Warning systems: ideas and examples for implementation” (2011). 36 sherri Lauver, “attracting and sustaining Youth participation in after school programs” (cambridge: harvard family research project, 2004).


center for american progress | positive student outcomes in community schools

The Center for American Progress is a nonpartisan research and educational institute dedicated to promoting a strong, just, and free America that ensures opportunity for all. We believe that Americans are bound together by a common commitment to these values and we aspire to ensure that our national policies reflect these values. We work to find progressive and pragmatic solutions to significant domestic and international problems and develop policy proposals that foster a government that is “of the people, by the people, and for the people.”

1333 H Street, NW, 10tH Floor, WaSHiNgtoN, DC 20005 • tel: 202-682-1611 • Fax: 202-682-1867 •

Sign up to vote on this title
UsefulNot useful